Thu, 18 December 2014
It’s safe to say that Docker has had a momentous year with the container-management startup gaining a lot of developer interest and scoring a lot of support from big tech companies like Amazon, Google, VMware and Microsoft.
Docker CEO Ben Golub came on to the Structure Show this week to talk about Docker’s year and what he envisions the company to be as it continues to grow (hint: it’s aiming for something similar to [company]VMware[/company]). Golub also talks about Docker’s raft of new orchestration features and shares his thoughts on the new CoreOS container technology and how that fits in with Docker.
In other news, Derrick Harris and Barb Darrow kick things off by looking at how Hortonworks and New Relic shares were holding up and the good news is -- they're doing pretty well at the ripe old age of 1 week.
Also on the docket, [company]IBM[/company] continues its cloud push by bringing a pantload of new data centers online -- in Frankfurt (for the all-important German market) as well as Mexico City and Tokyo. In October, IBM said it was working with local partner Tencent to add cloud services for the Chinese market, which reminds us that Amazon Web Services Beijing region remains in preview mode.
Hosts: Barbara Darrow, Derrick Harris and Jonathan Vanian
Guest: Ben Golub, CEO of Docker
Wed, 17 December 2014
Thu, 11 December 2014
If you want an informed opinion on the state of the cloud and the relative merits of the players, Sebastian Stadil's a good person to ask. Founder and CEO of Scalr, a multi-cloud management company, he keeps his finger on the pulse of all the players and -- perhaps more importantly -- their customers.
On this week's Structure Show he handicaps how [company]Google[/company] and [company]Microsoft[/company] are doing in public cloud not just technologically but in terms of their sales strategies which, when it comes to enterprise accounts, may be just as important as technology. And of course the company everyone is measuring by is [company]Amazon[/company] Web Services which leads the pack. Oh, and he's got lots to say about the OpenStack ecosystem as well; cloud technologies from [company]Oracle[/company], [company]Joyent[/company] and more.
It was a busy week in the funding arena with Nginx the web server company, DataGravity which makes "smart" storage arrays, Mesopshere infrastructure for scale out cloud all getting substantial VC rounds ($20 million, $50 million and $36 million respectively.) That ain't chicken feed.
This week's Hortonworks IPO puts the spotlight back on Hadoop and big data in a big way, which of course, gives us a chance to tout Gigaom's upcoming Structure Data event in New York with talks from Hortonworks CEO Rob Bearden, Cloudera CEO Tom Reilly and MapR CEO John Schroeder.
Wed, 10 December 2014
Direct download: Chrome_Show__Unlocking_Chromebooks_with_phones_and_watches.mp3
Category:Technology -- posted at: 2:41 PM
Thu, 4 December 2014
Alex Polvi seems a bit puzzled at the reaction to CoreOS's decision to develop its own basic container technology. Many saw that news, announced this week, as an aggressive move on Docker. It's fine with him that Docker wants to become a platform company and add management, orchestration, what-have-you to its eponymous Docker containers. His point is that many customers still need bare-bones containers and CoreOS wants to provide them.
On this week's Structure Show, Polvi agreed that Docker's move up the food chain from containers to platform is analogous to [company]VMware[/company]'s progression from a hypervisor company to virtualization-management company. If you ask today's startups if they'd like to be in VMware's shoes, most would say heck yeah, he said: There's nothing bad about being a platform company.
Still, it's indisputable that [company]CoreOS[/company]'s announcement struck a nerve with [company]Docker[/company] and ignited controversy. Polvi addresses lots of our questions here about the container ecosystem, where he sees CoreOS fitting in with Google's Kubernetes container management system and Mesos cluster management, and other topics including the state of the overall cloud computing ecosystem as he sees it..
It's a brisk and timely discussion so don't miss it.
In the first half of the show, Derrick Harris and I hash out the news around Hortonworks' new sub-$1 billion valuation; Amazon Web Services new no-money down instances, and Google's Automatic Statistician project.
Wed, 3 December 2014
Tue, 25 November 2014
Facebook has blazed a lot of trails by open sourcing some of its handiwork and driving the creation of the Open Compute Foundation to drive adoption of a standard servers for scale-out data centers. Now it's on to networking where it's talking up the new Data Center Fabric already rolled out in its Altoona, Iowa data center with broader deployment under way.
On this week's Structure Show, listen as Najam Ahmad, [company]Facebook[/company] director of network engineering patiently (and I mean patiently) explains what this technology means to application builders and why the move to more modular core-and-pod versus clustered systems is key to obliterating old networking bottlenecks. "For once, networking won't be the gating factor," he said.
Most simply put, in traditional networks, when the biggest and best top-of-the-line switches maxed out you had to reconfigure the whole cluster. With the fabric layout, bandwidth can be added much more easily and the network is limited -- in theory -- only by the size of the facility and available power. But he explains it much better than I ever could, so give it a listen.
Facebook's Look Back video extravaganza could have benefited from networking fabric and for more on how Facebook's infrastructure people got the plumbing together for 700 million videos in 3 weeks, check out a recent talk by Facebook capacity and performance engineer Goranka Bjedov linked in this story.
But first, first that we discuss the cloud opportunities and challenges [company]IBM[/company] faces and how the [company]Oracle[/company] cloud story just got a lot more interesting with its hiring of Mark Cavage, the brains behind Joyent's Manta distributed object store.
Thu, 20 November 2014
It's great that there are a ton of open source projects available for the taking, but checking them out and trying to get a feel for how they'll work out can be tricky for newbies (and even for some more grizzled veterans.)
That's why the [company]Netflix[/company] engineering brain trust rolled out ZeroToDocker last week --which promises to make it much easier for shop to test out say Asgard, Netflix's cloud deployment tool, in a limited setting before deploying it at scale. Basically, ZeroToDocker lets you (provided you have a [company]Docker[/company] host) run a single node of an NetflixOSS project with a one (1!) single command. Easy peasy.
"Once the images are downloaded it's Docker run, Docker run, Docker run with one parameter a piece and you don't need to understand much about it to get it running," Andrew Spyker, senior software engineer who joined Netflix from IBM not that long ago tells us.
Our other guest, Ruslan Meshenberg, director of cloud platform engineering also talks about why Netflix is all fired up about open-sourcing the technologies it builds to augment Amazon Web Services. Netflix is, after all, in the video streaming business.
Meshenberg acknowledges that Netflix OSS has no impact on the Netflix P&L but it is helpful all the same. "We want to validate what we're doing against the community. We don't want to be that unicorn stuck in a cul de sac, if you will. It's good to have external validation that we're not creating one-offs that we then have to maintain," he said.
And that OSS goodness makes it easier for Netflix to hire top engineering talent in the valley where it is a white-hot commodity.
Oh, and it also gives Netflix a leg up on hiring top engineering talent, which pretty much is all about open-source.
But before we get to all the Netflix/AWS goodness, Derrick Harris talks about all the action from [company]Google[/company], Stanford University, [company]Microsoft[/company] and others heating up the neural network sector; and we discuss the latest Microsoft Azure stumble. (Oh, since it's been a few sentences since I mentioned Docker, we'll also touch on the fact that people -- if they are so disposed -- can now run Docker instances from their Windows desktops.)
So get comfy and have a listen.
Wed, 19 November 2014
Fri, 14 November 2014
Amazon Web Services had a lot of stuff up its sleeve this week with its new MySQL-compatible Aurora relational database service, a new Docker container service, a service catalog, and Lambda, an event-driven compute service. This week on the show, Derrick Harris and I do our best to put all this into context.
But lest we forget, there is a world outside [company]Amazon[/company] and in that world HortonWorks filed for an initial public offering, which could show how well the public markets will grok Hadoop. New Relic also filed to go public.
And in other big news that would have been unthinkable a few years back, [company]Microsoft[/company], with help from [company]Xamarin[/company], is open-sourcing .Net trove of development tools and frameworks. I'm still not convinced the bulk of Linux and Mac developers were waiting for this to happen, but hey, it's a nice gesture no?
So get comfy and listen up. And we promise we'll have a guest next week.